Point Cloud Library (PCL): pcl::RegionGrowing< PointT, NormalT

Implements the well known Region Growing algorithm used for segmentation. More...

#include <pcl/segmentation/region_growing.h>

Inheritance diagram for pcl::RegionGrowing< PointT, NormalT >:

Public Types
using	KdTree = pcl::search::Search< PointT >

using	KdTreePtr = typename KdTree::Ptr

using	Normal = pcl::PointCloud< NormalT >

using	NormalPtr = typename Normal::Ptr

using	PointCloud = pcl::PointCloud< PointT >

Public Types inherited from pcl::PCLBase< PointT >
using	PointCloud = pcl::PointCloud< PointT >

using	PointCloudPtr = typename PointCloud::Ptr

using	PointCloudConstPtr = typename PointCloud::ConstPtr

using	PointIndicesPtr = PointIndices::Ptr

using	PointIndicesConstPtr = PointIndices::ConstPtr

Public Member Functions
	RegionGrowing ()
	Constructor that sets default values for member variables. More...

	~RegionGrowing ()
	This destructor destroys the cloud, normals and search method used for finding KNN. More...

int	getMinClusterSize ()
	Get the minimum number of points that a cluster needs to contain in order to be considered valid. More...

void	setMinClusterSize (int min_cluster_size)
	Set the minimum number of points that a cluster needs to contain in order to be considered valid. More...

int	getMaxClusterSize ()
	Get the maximum number of points that a cluster needs to contain in order to be considered valid. More...

void	setMaxClusterSize (int max_cluster_size)
	Set the maximum number of points that a cluster needs to contain in order to be considered valid. More...

bool	getSmoothModeFlag () const
	Returns the flag value. More...

void	setSmoothModeFlag (bool value)
	This function allows to turn on/off the smoothness constraint. More...

bool	getCurvatureTestFlag () const
	Returns the flag that signalize if the curvature test is turned on/off. More...

virtual void	setCurvatureTestFlag (bool value)
	Allows to turn on/off the curvature test. More...

bool	getResidualTestFlag () const
	Returns the flag that signalize if the residual test is turned on/off. More...

virtual void	setResidualTestFlag (bool value)
	Allows to turn on/off the residual test. More...

float	getSmoothnessThreshold () const
	Returns smoothness threshold. More...

void	setSmoothnessThreshold (float theta)
	Allows to set smoothness threshold used for testing the points. More...

float	getResidualThreshold () const
	Returns residual threshold. More...

void	setResidualThreshold (float residual)
	Allows to set residual threshold used for testing the points. More...

float	getCurvatureThreshold () const
	Returns curvature threshold. More...

void	setCurvatureThreshold (float curvature)
	Allows to set curvature threshold used for testing the points. More...

unsigned int	getNumberOfNeighbours () const
	Returns the number of nearest neighbours used for KNN. More...

void	setNumberOfNeighbours (unsigned int neighbour_number)
	Allows to set the number of neighbours. More...

KdTreePtr	getSearchMethod () const
	Returns the pointer to the search method that is used for KNN. More...

void	setSearchMethod (const KdTreePtr &tree)
	Allows to set search method that will be used for finding KNN. More...

NormalPtr	getInputNormals () const
	Returns normals. More...

void	setInputNormals (const NormalPtr &norm)
	This method sets the normals. More...

virtual void	extract (std::vector< pcl::PointIndices > &clusters)
	This method launches the segmentation algorithm and returns the clusters that were obtained during the segmentation. More...

virtual void	getSegmentFromPoint (int index, pcl::PointIndices &cluster)
	For a given point this function builds a segment to which it belongs and returns this segment. More...

pcl::PointCloud< pcl::PointXYZRGB >::Ptr	getColoredCloud ()
	If the cloud was successfully segmented, then function returns colored cloud. More...

pcl::PointCloud< pcl::PointXYZRGBA >::Ptr	getColoredCloudRGBA ()
	If the cloud was successfully segmented, then function returns colored cloud. More...

Public Member Functions inherited from pcl::PCLBase< PointT >
	PCLBase ()
	Empty constructor. More...

	PCLBase (const PCLBase &base)
	Copy constructor. More...

virtual	~PCLBase ()=default
	Destructor. More...

virtual void	setInputCloud (const PointCloudConstPtr &cloud)
	Provide a pointer to the input dataset. More...

const PointCloudConstPtr	getInputCloud () const
	Get a pointer to the input point cloud dataset. More...

virtual void	setIndices (const IndicesPtr &indices)
	Provide a pointer to the vector of indices that represents the input data. More...

virtual void	setIndices (const IndicesConstPtr &indices)
	Provide a pointer to the vector of indices that represents the input data. More...

virtual void	setIndices (const PointIndicesConstPtr &indices)
	Provide a pointer to the vector of indices that represents the input data. More...

virtual void	setIndices (std::size_t row_start, std::size_t col_start, std::size_t nb_rows, std::size_t nb_cols)
	Set the indices for the points laying within an interest region of the point cloud. More...

IndicesPtr	getIndices ()
	Get a pointer to the vector of indices used. More...

const IndicesConstPtr	getIndices () const
	Get a pointer to the vector of indices used. More...

const PointT &	operator[] (std::size_t pos) const
	Override PointCloud operator[] to shorten code. More...

Protected Member Functions
virtual bool	prepareForSegmentation ()
	This method simply checks if it is possible to execute the segmentation algorithm with the current settings. More...

virtual void	findPointNeighbours ()
	This method finds KNN for each point and saves them to the array because the algorithm needs to find KNN a few times. More...

void	applySmoothRegionGrowingAlgorithm ()
	This function implements the algorithm described in the article "Segmentation of point clouds using smoothness constraint" by T. More...

int	growRegion (int initial_seed, int segment_number)
	This method grows a segment for the given seed point. More...

virtual bool	validatePoint (int initial_seed, int point, int nghbr, bool &is_a_seed) const
	This function is checking if the point with index 'nghbr' belongs to the segment. More...

void	assembleRegions ()
	This function simply assembles the regions from list of point labels. More...

Protected Member Functions inherited from pcl::PCLBase< PointT >
bool	initCompute ()
	This method should get called before starting the actual computation. More...

bool	deinitCompute ()
	This method should get called after finishing the actual computation. More...

Protected Attributes
int	min_pts_per_cluster_
	Stores the minimum number of points that a cluster needs to contain in order to be considered valid. More...

int	max_pts_per_cluster_
	Stores the maximum number of points that a cluster needs to contain in order to be considered valid. More...

bool	smooth_mode_flag_
	Flag that signalizes if the smoothness constraint will be used. More...

bool	curvature_flag_
	If set to true then curvature test will be done during segmentation. More...

bool	residual_flag_
	If set to true then residual test will be done during segmentation. More...

float	theta_threshold_
	Thershold used for testing the smoothness between points. More...

float	residual_threshold_
	Thershold used in residual test. More...

float	curvature_threshold_
	Thershold used in curvature test. More...

unsigned int	neighbour_number_
	Number of neighbours to find. More...

KdTreePtr	search_
	Serch method that will be used for KNN. More...

NormalPtr	normals_
	Contains normals of the points that will be segmented. More...

std::vector< std::vector< int > >	point_neighbours_
	Contains neighbours of each point. More...

std::vector< int >	point_labels_
	Point labels that tells to which segment each point belongs. More...

bool	normal_flag_
	If set to true then normal/smoothness test will be done during segmentation. More...

std::vector< int >	num_pts_in_segment_
	Tells how much points each segment contains. More...

std::vector< pcl::PointIndices >	clusters_
	After the segmentation it will contain the segments. More...

int	number_of_segments_
	Stores the number of segments. More...

Protected Attributes inherited from pcl::PCLBase< PointT >
PointCloudConstPtr	input_
	The input point cloud dataset. More...

IndicesPtr	indices_
	A pointer to the vector of point indices to use. More...

bool	use_indices_
	Set to true if point indices are used. More...

bool	fake_indices_
	If no set of indices are given, we construct a set of fake indices that mimic the input PointCloud. More...

Detailed Description

template<typename PointT, typename NormalT>
class pcl::RegionGrowing< PointT, NormalT >

Implements the well known Region Growing algorithm used for segmentation.

Description can be found in the article "Segmentation of point clouds using smoothness constraint" by T. Rabbania, F. A. van den Heuvelb, G. Vosselmanc. In addition to residual test, the possibility to test curvature is added.

Definition at line 60 of file region_growing.h.

Member Typedef Documentation

◆ KdTree

template<typename PointT , typename NormalT >

using pcl::RegionGrowing< PointT, NormalT >::KdTree = pcl::search::Search<PointT>

Definition at line 64 of file region_growing.h.

◆ KdTreePtr

template<typename PointT , typename NormalT >

using pcl::RegionGrowing< PointT, NormalT >::KdTreePtr = typename KdTree::Ptr

Definition at line 65 of file region_growing.h.

◆ Normal

template<typename PointT , typename NormalT >

using pcl::RegionGrowing< PointT, NormalT >::Normal = pcl::PointCloud<NormalT>

Definition at line 66 of file region_growing.h.

◆ NormalPtr

template<typename PointT , typename NormalT >

using pcl::RegionGrowing< PointT, NormalT >::NormalPtr = typename Normal::Ptr

Definition at line 67 of file region_growing.h.

◆ PointCloud

template<typename PointT , typename NormalT >

using pcl::RegionGrowing< PointT, NormalT >::PointCloud = pcl::PointCloud<PointT>

Definition at line 68 of file region_growing.h.

Constructor & Destructor Documentation

◆ RegionGrowing()

template<typename PointT , typename NormalT >

pcl::RegionGrowing< PointT, NormalT >::RegionGrowing

Constructor that sets default values for member variables.

Definition at line 57 of file region_growing.hpp.

◆ ~RegionGrowing()

template<typename PointT , typename NormalT >

pcl::RegionGrowing< PointT, NormalT >::~RegionGrowing

This destructor destroys the cloud, normals and search method used for finding KNN.

In other words it frees memory.

Definition at line 80 of file region_growing.hpp.

Member Function Documentation

◆ applySmoothRegionGrowingAlgorithm()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::applySmoothRegionGrowingAlgorithm

protected

This function implements the algorithm described in the article "Segmentation of point clouds using smoothness constraint" by T.

Rabbania, F. A. van den Heuvelb, G. Vosselmanc.

Definition at line 376 of file region_growing.hpp.

◆ assembleRegions()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::assembleRegions

protected

This function simply assembles the regions from list of point labels.

Each cluster is an array of point indices.

Definition at line 539 of file region_growing.hpp.

Referenced by pcl::RegionGrowingRGB< PointT, NormalT >::extract(), and pcl::RegionGrowingRGB< PointT, NormalT >::getSegmentFromPoint().

◆ extract()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::extract ( std::vector< pcl::PointIndices > & clusters )

virtual

This method launches the segmentation algorithm and returns the clusters that were obtained during the segmentation.

Parameters

[out] clusters clusters that were obtained. Each cluster is an array of point indices.

Reimplemented in pcl::RegionGrowingRGB< PointT, NormalT >.

Definition at line 250 of file region_growing.hpp.

◆ findPointNeighbours()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::findPointNeighbours

protectedvirtual

This method finds KNN for each point and saves them to the array because the algorithm needs to find KNN a few times.

Reimplemented in pcl::RegionGrowingRGB< PointT, NormalT >.

Definition at line 343 of file region_growing.hpp.

◆ getColoredCloud()

template<typename PointT , typename NormalT >

pcl::PointCloud< pcl::PointXYZRGB >::Ptr pcl::RegionGrowing< PointT, NormalT >::getColoredCloud

If the cloud was successfully segmented, then function returns colored cloud.

Otherwise it returns an empty pointer. Points that belong to the same segment have the same color. But this function doesn't guarantee that different segments will have different color(it all depends on RNG). Points that were not listed in the indices array will have red color.

Definition at line 639 of file region_growing.hpp.

◆ getColoredCloudRGBA()

template<typename PointT , typename NormalT >

pcl::PointCloud< pcl::PointXYZRGBA >::Ptr pcl::RegionGrowing< PointT, NormalT >::getColoredCloudRGBA

If the cloud was successfully segmented, then function returns colored cloud.

Definition at line 691 of file region_growing.hpp.

◆ getCurvatureTestFlag()

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::getCurvatureTestFlag

Returns the flag that signalize if the curvature test is turned on/off.

Definition at line 132 of file region_growing.hpp.

◆ getCurvatureThreshold()

template<typename PointT , typename NormalT >

float pcl::RegionGrowing< PointT, NormalT >::getCurvatureThreshold

Returns curvature threshold.

Definition at line 194 of file region_growing.hpp.

◆ getInputNormals()

template<typename PointT , typename NormalT >

pcl::RegionGrowing< PointT, NormalT >::NormalPtr pcl::RegionGrowing< PointT, NormalT >::getInputNormals

Returns normals.

Definition at line 236 of file region_growing.hpp.

◆ getMaxClusterSize()

template<typename PointT , typename NormalT >

int pcl::RegionGrowing< PointT, NormalT >::getMaxClusterSize

Get the maximum number of points that a cluster needs to contain in order to be considered valid.

Definition at line 104 of file region_growing.hpp.

◆ getMinClusterSize()

template<typename PointT , typename NormalT >

int pcl::RegionGrowing< PointT, NormalT >::getMinClusterSize

Get the minimum number of points that a cluster needs to contain in order to be considered valid.

Definition at line 90 of file region_growing.hpp.

◆ getNumberOfNeighbours()

template<typename PointT , typename NormalT >

unsigned int pcl::RegionGrowing< PointT, NormalT >::getNumberOfNeighbours

Returns the number of nearest neighbours used for KNN.

Definition at line 208 of file region_growing.hpp.

◆ getResidualTestFlag()

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::getResidualTestFlag

Returns the flag that signalize if the residual test is turned on/off.

Definition at line 149 of file region_growing.hpp.

◆ getResidualThreshold()

template<typename PointT , typename NormalT >

float pcl::RegionGrowing< PointT, NormalT >::getResidualThreshold

Returns residual threshold.

Definition at line 180 of file region_growing.hpp.

◆ getSearchMethod()

template<typename PointT , typename NormalT >

pcl::RegionGrowing< PointT, NormalT >::KdTreePtr pcl::RegionGrowing< PointT, NormalT >::getSearchMethod

Returns the pointer to the search method that is used for KNN.

Definition at line 222 of file region_growing.hpp.

◆ getSegmentFromPoint()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::getSegmentFromPoint	(	int	index,
		pcl::PointIndices &	cluster
	)

virtual

For a given point this function builds a segment to which it belongs and returns this segment.

Parameters

[in]	index	index of the initial point which will be the seed for growing a segment.
[out]	cluster	cluster to which the point belongs.

Definition at line 570 of file region_growing.hpp.

◆ getSmoothModeFlag()

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::getSmoothModeFlag

Returns the flag value.

This flag signalizes which mode of algorithm will be used. If it is set to true than it will work as said in the article. This means that it will be testing the angle between normal of the current point and it's neighbours normal. Otherwise, it will be testing the angle between normal of the current point and normal of the initial point that was chosen for growing new segment.

Definition at line 118 of file region_growing.hpp.

◆ getSmoothnessThreshold()

template<typename PointT , typename NormalT >

float pcl::RegionGrowing< PointT, NormalT >::getSmoothnessThreshold

Returns smoothness threshold.

Definition at line 166 of file region_growing.hpp.

◆ growRegion()

template<typename PointT , typename NormalT >

int pcl::RegionGrowing< PointT, NormalT >::growRegion	(	int	initial_seed,
		int	segment_number
	)

protected

This method grows a segment for the given seed point.

And returns the number of its points.

Parameters

[in]	initial_seed	index of the point that will serve as the seed point
[in]	segment_number	indicates which number this segment will have

Definition at line 433 of file region_growing.hpp.

◆ prepareForSegmentation()

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::prepareForSegmentation

protectedvirtual

This method simply checks if it is possible to execute the segmentation algorithm with the current settings.

If it is possible then it returns true.

Reimplemented in pcl::RegionGrowingRGB< PointT, NormalT >.

Definition at line 297 of file region_growing.hpp.

◆ setCurvatureTestFlag()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setCurvatureTestFlag ( bool value )

virtual

Allows to turn on/off the curvature test.

Note that at least one test (residual or curvature) must be turned on. If you are turning curvature test off then residual test will be turned on automatically.

Parameters

[in] value new value for curvature test. If set to true then the test will be turned on

Reimplemented in pcl::RegionGrowingRGB< PointT, NormalT >.

Definition at line 139 of file region_growing.hpp.

◆ setCurvatureThreshold()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setCurvatureThreshold ( float curvature )

Allows to set curvature threshold used for testing the points.

Parameters

[in] curvature new threshold value for curvature testing

Definition at line 201 of file region_growing.hpp.

◆ setInputNormals()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setInputNormals ( const NormalPtr & norm )

This method sets the normals.

They are needed for the algorithm, so if no normals will be set, the algorithm would not be able to segment the points.

Parameters

[in] norm normals that will be used in the algorithm

Definition at line 243 of file region_growing.hpp.

◆ setMaxClusterSize()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setMaxClusterSize ( int max_cluster_size )

Set the maximum number of points that a cluster needs to contain in order to be considered valid.

Definition at line 111 of file region_growing.hpp.

◆ setMinClusterSize()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setMinClusterSize ( int min_cluster_size )

Set the minimum number of points that a cluster needs to contain in order to be considered valid.

Definition at line 97 of file region_growing.hpp.

◆ setNumberOfNeighbours()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setNumberOfNeighbours ( unsigned int neighbour_number )

Allows to set the number of neighbours.

For more information check the article.

Parameters

[in] neighbour_number number of neighbours to use

Definition at line 215 of file region_growing.hpp.

◆ setResidualTestFlag()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setResidualTestFlag ( bool value )

virtual

Allows to turn on/off the residual test.

Note that at least one test (residual or curvature) must be turned on. If you are turning residual test off then curvature test will be turned on automatically.

Parameters

[in] value new value for residual test. If set to true then the test will be turned on

Reimplemented in pcl::RegionGrowingRGB< PointT, NormalT >.

Definition at line 156 of file region_growing.hpp.

◆ setResidualThreshold()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setResidualThreshold ( float residual )

Allows to set residual threshold used for testing the points.

Parameters

[in] residual new threshold value for residual testing

Definition at line 187 of file region_growing.hpp.

◆ setSearchMethod()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setSearchMethod ( const KdTreePtr & tree )

Allows to set search method that will be used for finding KNN.

Parameters

[in] tree pointer to a KdTree

Definition at line 229 of file region_growing.hpp.

◆ setSmoothModeFlag()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setSmoothModeFlag ( bool value )

This function allows to turn on/off the smoothness constraint.

Parameters

[in] value new mode value, if set to true then the smooth version will be used.

Definition at line 125 of file region_growing.hpp.

◆ setSmoothnessThreshold()

template<typename PointT , typename NormalT >

void pcl::RegionGrowing< PointT, NormalT >::setSmoothnessThreshold ( float theta )

Allows to set smoothness threshold used for testing the points.

Parameters

[in] theta new threshold value for the angle between normals

Definition at line 173 of file region_growing.hpp.

◆ validatePoint()

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::validatePoint	(	int	initial_seed,
		int	point,
		int	nghbr,
		bool &	is_a_seed
	)		const

protectedvirtual

This function is checking if the point with index 'nghbr' belongs to the segment.

If so, then it returns true. It also checks if this point can serve as the seed.

Parameters

[in]	initial_seed	index of the initial point that was passed to the growRegion() function
[in]	point	index of the current seed point
[in]	nghbr	index of the point that is neighbour of the current seed
[out]	is_a_seed	this value is set to true if the point with index 'nghbr' can serve as the seed

Definition at line 483 of file region_growing.hpp.

Member Data Documentation

◆ clusters_

template<typename PointT , typename NormalT >

std::vector<pcl::PointIndices> pcl::RegionGrowing< PointT, NormalT >::clusters_

protected

After the segmentation it will contain the segments.

Definition at line 329 of file region_growing.h.

◆ curvature_flag_

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::curvature_flag_

protected

If set to true then curvature test will be done during segmentation.

Definition at line 291 of file region_growing.h.

◆ curvature_threshold_

template<typename PointT , typename NormalT >

float pcl::RegionGrowing< PointT, NormalT >::curvature_threshold_

protected

Thershold used in curvature test.

Definition at line 303 of file region_growing.h.

◆ max_pts_per_cluster_

template<typename PointT , typename NormalT >

int pcl::RegionGrowing< PointT, NormalT >::max_pts_per_cluster_

protected

Stores the maximum number of points that a cluster needs to contain in order to be considered valid.

Definition at line 285 of file region_growing.h.

◆ min_pts_per_cluster_

template<typename PointT , typename NormalT >

int pcl::RegionGrowing< PointT, NormalT >::min_pts_per_cluster_

protected

Stores the minimum number of points that a cluster needs to contain in order to be considered valid.

Definition at line 282 of file region_growing.h.

◆ neighbour_number_

template<typename PointT , typename NormalT >

unsigned int pcl::RegionGrowing< PointT, NormalT >::neighbour_number_

protected

Number of neighbours to find.

Definition at line 306 of file region_growing.h.

◆ normal_flag_

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::normal_flag_

protected

If set to true then normal/smoothness test will be done during segmentation.

It is always set to true for the usual region growing algorithm. It is used for turning on/off the test for smoothness in the child class RegionGrowingRGB.

Definition at line 323 of file region_growing.h.

◆ normals_

template<typename PointT , typename NormalT >

NormalPtr pcl::RegionGrowing< PointT, NormalT >::normals_

protected

Contains normals of the points that will be segmented.

Definition at line 312 of file region_growing.h.

◆ num_pts_in_segment_

template<typename PointT , typename NormalT >

std::vector<int> pcl::RegionGrowing< PointT, NormalT >::num_pts_in_segment_

protected

Tells how much points each segment contains.

Used for reserving memory.

Definition at line 326 of file region_growing.h.

◆ number_of_segments_

template<typename PointT , typename NormalT >

int pcl::RegionGrowing< PointT, NormalT >::number_of_segments_

protected

Stores the number of segments.

Definition at line 332 of file region_growing.h.

◆ point_labels_

template<typename PointT , typename NormalT >

std::vector<int> pcl::RegionGrowing< PointT, NormalT >::point_labels_

protected

Point labels that tells to which segment each point belongs.

Definition at line 318 of file region_growing.h.

◆ point_neighbours_

template<typename PointT , typename NormalT >

std::vector<std::vector<int> > pcl::RegionGrowing< PointT, NormalT >::point_neighbours_

protected

Contains neighbours of each point.

Definition at line 315 of file region_growing.h.

◆ residual_flag_

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::residual_flag_

protected

If set to true then residual test will be done during segmentation.

Definition at line 294 of file region_growing.h.

◆ residual_threshold_

template<typename PointT , typename NormalT >

float pcl::RegionGrowing< PointT, NormalT >::residual_threshold_

protected

Thershold used in residual test.

Definition at line 300 of file region_growing.h.

◆ search_

template<typename PointT , typename NormalT >

KdTreePtr pcl::RegionGrowing< PointT, NormalT >::search_

protected

Serch method that will be used for KNN.

Definition at line 309 of file region_growing.h.

◆ smooth_mode_flag_

template<typename PointT , typename NormalT >

bool pcl::RegionGrowing< PointT, NormalT >::smooth_mode_flag_

protected

Flag that signalizes if the smoothness constraint will be used.

Definition at line 288 of file region_growing.h.

◆ theta_threshold_

template<typename PointT , typename NormalT >

float pcl::RegionGrowing< PointT, NormalT >::theta_threshold_

protected

Thershold used for testing the smoothness between points.

Definition at line 297 of file region_growing.h.

The documentation for this class was generated from the following files:

pcl/segmentation/region_growing.h
pcl/segmentation/impl/region_growing.hpp

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

template<typename PointT, typename NormalT> class pcl::RegionGrowing< PointT, NormalT >

Member Typedef Documentation

◆ KdTree

◆ KdTreePtr

◆ Normal

◆ NormalPtr

◆ PointCloud

Constructor & Destructor Documentation

◆ RegionGrowing()

◆ ~RegionGrowing()

Member Function Documentation

◆ applySmoothRegionGrowingAlgorithm()

◆ assembleRegions()

◆ extract()

◆ findPointNeighbours()

◆ getColoredCloud()

◆ getColoredCloudRGBA()

◆ getCurvatureTestFlag()

◆ getCurvatureThreshold()

◆ getInputNormals()

◆ getMaxClusterSize()

◆ getMinClusterSize()

◆ getNumberOfNeighbours()

◆ getResidualTestFlag()

◆ getResidualThreshold()

◆ getSearchMethod()

◆ getSegmentFromPoint()

◆ getSmoothModeFlag()

◆ getSmoothnessThreshold()

◆ growRegion()

◆ prepareForSegmentation()

◆ setCurvatureTestFlag()

◆ setCurvatureThreshold()

◆ setInputNormals()

◆ setMaxClusterSize()

◆ setMinClusterSize()

◆ setNumberOfNeighbours()

◆ setResidualTestFlag()

◆ setResidualThreshold()

◆ setSearchMethod()

◆ setSmoothModeFlag()

◆ setSmoothnessThreshold()

◆ validatePoint()

Member Data Documentation

◆ clusters_

◆ curvature_flag_

◆ curvature_threshold_

◆ max_pts_per_cluster_

◆ min_pts_per_cluster_

◆ neighbour_number_

◆ normal_flag_

◆ normals_

◆ num_pts_in_segment_

◆ number_of_segments_

◆ point_labels_

◆ point_neighbours_

◆ residual_flag_

◆ residual_threshold_

◆ search_

◆ smooth_mode_flag_

◆ theta_threshold_

template<typename PointT, typename NormalT>
class pcl::RegionGrowing< PointT, NormalT >